Mapping services providers face significant technical challenges to providing map data that accurately reflect real-word points-of-interests or features. For example, providing data on locations where vehicles can park, stop, or stand (e.g., parking lots, loading zones, standing zones, alleys, etc.) when they arrive at their destinations is an area of interest. Mapping such locations, however, has traditionally been a resource-intensive and time-consuming manual process, which can limit how often map services update their map data and related products.